Adversaries may steal data by exfiltrating it over a different protocol than that of the existing command and control channel. The data may also be sent to an alternate network location from the main command and control server.
Alternate protocols include FTP, SMTP, HTTP/S, DNS, SMB, or any other network protocol not being used as the main command and control channel. Adversaries may also opt to encrypt and/or obfuscate these alternate channels.
Exfiltration Over Alternative Protocol can be done using various common operating system utilities such as Net/SMB or FTP.[1] On macOS and Linux curl may be used to invoke protocols such as HTTP/S or FTP/S to exfiltrate data from a system.[2]
Many IaaS and SaaS platforms (such as Microsoft Exchange, Microsoft SharePoint, GitHub, and AWS S3) support the direct download of files, emails, source code, and other sensitive information via the web console or Cloud API.
| 效应类型 | 是否存在 |
|---|---|
| 特征伪装 | ✅ |
| 行为透明 | ❌ |
| 数据遮蔽 | ✅ |
| 时空释痕 | ✅ |
攻击者通过深度协议模拟和业务流量克隆，使渗出流量在协议交互模式、数据包结构、时序特征等维度与合法流量高度一致。例如将渗出数据封装为视频流协议的有效载荷，或模仿云服务API的标准请求格式，使得传统基于协议指纹或流量特征的检测手段无法有效区分恶意渗出与正常业务交互，实现攻击行为的特征层面隐匿。
攻击者在渗出过程中普遍采用分层加密与协议嵌套技术，对窃取数据实施传输层与应用层的双重加密保护。例如在HTTP/2流中嵌套使用TLS 1.3加密通道传输AES-GCM加密的渗出数据，或在DNS渗出中通过DNSSEC扩展对数据载荷进行签名加密。此类手法使得即便防御方截获渗出流量，仍需突破多层加密才能解析原始数据，显著提高攻击行为的反向分析难度。
通过低频分片与多协议交替传输策略，攻击者将集中式数据渗出行为分散为长时间跨度的多个微流量事件。例如将10GB数据分割为10KB片段，通过每日单次HTTP POST请求持续渗出300天，每个请求的流量规模均低于网络监控系统的异常阈值。这种时空维度上的威胁特征稀释，使得传统基于实时流量分析或短期日志审计的检测机制难以捕捉完整攻击链，必须依赖跨月级的数据聚合分析才能识别威胁，极大提升防御成本。
| ID | Name | Description |
|---|---|---|
| S0677 | AADInternals |
AADInternals can directly download cloud user data such as OneDrive files.[3] |
| S0482 | Bundlore |
Bundlore uses the |
| S0631 | Chaes |
Chaes has exfiltrated its collected data from the infected machine to the C2, sometimes using the MIME protocol.[4] |
| S0503 | FrameworkPOS |
FrameworkPOS can use DNS tunneling for exfiltration of credit card data.[5] |
| S0203 | Hydraq |
Hydraq connects to a predefined domain on port 443 to exfil gathered information.[6] |
| S0641 | Kobalos |
Kobalos can exfiltrate credentials over the network via UDP.[7] |
| G1040 | Play |
Play has used WinSCP to exfiltrate data to actor-controlled accounts.[8][9] |
| S0428 | PoetRAT |
PoetRAT has used a .NET tool named dog.exe to exiltrate information over an e-mail account.[10] |
| G0139 | TeamTNT |
TeamTNT has sent locally staged files with collected credentials to C2 servers using cURL.[11] |
| ID | Mitigation | Description |
|---|---|---|
| M1057 | Data Loss Prevention |
Data loss prevention can detect and block sensitive data being uploaded via web browsers. |
| M1037 | Filter Network Traffic |
Enforce proxies and use dedicated servers for services such as DNS and only allow those systems to communicate over respective ports/protocols, instead of all systems within a network. Cloud service providers support IP-based restrictions when accessing cloud resources. Consider using IP allowlisting along with user account management to ensure that data access is restricted not only to valid users but only from expected IP ranges to mitigate the use of stolen credentials to access data. |
| M1031 | Network Intrusion Prevention |
Network intrusion detection and prevention systems that use network signatures to identify traffic for specific adversary command and control infrastructure and malware can be used to mitigate activity at the network level. |
| M1030 | Network Segmentation |
Follow best practices for network firewall configurations to allow only necessary ports and traffic to enter and exit the network.[12] |
| M1022 | Restrict File and Directory Permissions |
Use access control lists on cloud storage systems and objects. |
| M1018 | User Account Management |
Configure user permissions groups and roles for access to cloud storage.[13] Implement strict Identity and Access Management (IAM) controls to prevent access to storage solutions except for the applications, users, and services that require access.[14] Ensure that temporary access tokens are issued rather than permanent credentials, especially when access is being granted to entities outside of the internal security boundary.[15] |
| ID | Data Source | Data Component | Detects |
|---|---|---|---|
| DS0015 | Application Log | Application Log Content |
Monitor cloud-based file hosting services, such as Google Drive and Microsoft OneDrive, for unusual instances of file downloads – for example, many downloads by a single user in a short period of time. In environments with high-maturity, it may be possible to leverage User-Behavioral Analytics (UBA) platforms to detect and alert on user-based anomalies. Additionally, data loss prevention policies can be defined to detect and alert on exfiltration events on particularly sensitive data. |
| DS0010 | Cloud Storage | Cloud Storage Access |
Monitor for unusual queries to the cloud provider's storage service. Activity originating from unexpected sources may indicate improper permissions are set and are allowing access to data. Additionally, detecting failed attempts by a user for a certain object, followed by escalation of privileges by the same user, and access to the same object may be an indication of suspicious activity. |
| DS0017 | Command | Command Execution |
Monitor executed commands and arguments that may steal data by exfiltrating it over a different protocol than that of the existing command and control channel. |
| DS0022 | File | File Access |
Monitor for suspicious files (i.e. .pdf, .docx, .jpg, etc.) viewed in isolation that may steal data by exfiltrating it over a different protocol than that of the existing command and control channel. |
| DS0029 | Network Traffic | Network Connection Creation |
Monitor for newly constructed network connections that are sent or received by untrusted hosts. |
| Network Traffic Content |
Monitor and analyze traffic patterns and packet inspection associated to protocol(s) that do not follow the expected protocol standards and traffic flows (e.g extraneous packets that do not belong to established flows, gratuitous or anomalous traffic patterns, anomalous syntax, or structure). Consider correlation with process monitoring and command line to detect anomalous processes execution and command line arguments associated to traffic patterns (e.g. monitor anomalies in use of files that do not normally initiate connections for respective protocol(s)). |
||
| Network Traffic Flow |
Monitor network data for uncommon data flows. Processes utilizing the network that do not normally have network communication or have never been seen before are suspicious. |